Deduplication systems perform backup processes for amounts of data that are continually growing in the information age. Storage capacity requirements for deduplication systems are increasing at a rapid rate. Yet, achieving scalable storage management remains a challenge, especially as deduplication systems approach and surpass the petabyte (PB) level. File systems, such as Veritas File System (VxFS) and XFS may only efficiently manage a limited amount of disk or other storage space. Load balancing also becomes problematic with large amounts of data. The efficiency of data writes and reference updates may decrease with large amounts of data and many clients sending data under different policies to a deduplication system. It is within this context that the embodiments arise.